speecht5_finetuned_voxpopuli_nl

This model is a fine-tuned version of AFZAL0008/speecht5_finetuned_voxpopuli_nl on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.7208

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 1e-05
• train_batch_size: 4
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 5
• training_steps: 100
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
7.4281	0.0232	10	0.8151
7.1065	0.0464	20	0.7918
6.9129	0.0696	30	0.7637
6.9501	0.0928	40	0.7524
6.9925	0.1160	50	0.7506
6.3808	0.1393	60	0.7367
6.5429	0.1625	70	0.7336
6.5886	0.1857	80	0.7292
6.4179	0.2089	90	0.7222
6.4979	0.2321	100	0.7208

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0